AI + Digital

AI changed discovery.
It didn’t change the fundamentals.

Websites. AEO. Products. Digital foundations.

Search is changing quickly. Google, ChatGPT, Perplexity and other AI systems are reshaping how people discover businesses, but the sites that win still need to be credible, understandable, and useful.

I think about digital foundations through three things: credible, crawlable, convertible.

How people find you now

  • Google Search
  • ChatGPT AI answers
  • Perplexity Research
  • Gemini AI search
  • YouTube Videos
  • Reddit Community
yourbusiness.com
Your business Website · Content · Reputation
A brand
people trust
  • Credible People can trust what they see.
  • Crawlable Easy to understand for search and AI.
  • Convertible Turns visitors into customers.
Different discovery engines. Same foundation.

How I got here

I didn’t start with AI.

I started by trying to make stores convert better. That pulled me deeper into websites, structure, content, search, and eventually AI discovery.

Each layer only made sense because of the one underneath it.

  1. 01

    E-commerce

    Selling real products and learning what actually makes someone buy.

  2. 02

    Websites

    Conversion problems are usually structure problems wearing a design costume.

  3. 03

    Search + structure

    Hierarchy, internal links and schema — making a business legible to machines.

  4. 04

    AEO

    The same fundamentals, now read by answer engines instead of only ranked pages.

  5. 05

    Products + apps

    Sometimes the answer is not a better page. It is a better tool.

Framework

Credible. Crawlable. Convertible.

Whether it’s a website, an AI search strategy, or a digital product, I keep coming back to the same three things.

  1. 01

    Credible

    Can a human or an AI system trust what it sees?

    • Authority in a real category
    • Proof you can point at
    • Clear, specific information
    • Consistency across every surface
    • Reviews and third-party mentions
    • First-party content nobody else has
  2. 02

    Crawlable

    Can search engines and AI systems actually understand the business?

    • Site architecture that matches reality
    • Internal linking with intent
    • Structured data that describes it
    • Semantic content, not keyword filler
    • Indexability and technical health
    • Entity clarity — who this is, exactly
  3. 03

    Convertible

    Once someone gets there, does the experience actually work?

    • UX that respects why they came
    • Messaging that answers, not announces
    • Mobile as the primary case
    • Page structure you can skim
    • Trust and speed, in that order
    • One obvious next action

AEO

AEO is simpler than most marketers make it sound.

AI systems still need what search engines and users have always needed: clear structure, useful content, trustworthy signals, and strong answers.

  1. 01 Foundation

    Nothing works until a system can reach the site and tell what it is.

    • Technical health
    • Crawlability
    • Schema
    • Entity clarity
  2. 02 Authority

    Being findable is not the same as being worth citing.

    • First-party content
    • Third-party mentions
    • Reviews
    • Citations
  3. 03 Coverage

    If a question has no page, there is nothing to answer with.

    • Service coverage
    • Product coverage
    • Location coverage
    • Question coverage
  4. 04 Measurement

    Visibility you cannot see is indistinguishable from a guess.

    • Prompt tracking
    • AI visibility
    • Search visibility
    • Leads and conversions

Discovery today

Discovery is no longer one path.

People find a business in a dozen places now, and most of those places are reading the same source material.

  • Google
  • Google AI overviews
  • ChatGPT
  • Perplexity
  • Reddit
  • YouTube
  • Marketplaces
  • Social

The information layer

Your website is no longer just the destination.

It is part of the information layer these systems use to understand your business — quoted, summarised and compared before anyone clicks anything.

  • Pages and hierarchy
  • Structured data
  • Reviews and mentions
  • Entity signals

Websites

The website is still the foundation.

A beautiful site that can’t be understood is weak. A technically clean site that doesn’t convert is also weak. The best sites do both, and the work is mostly in the seams between them.

This is where the e-commerce years actually pay off — I learned all of this by watching real people fail to buy things.

  • Hierarchy

    What matters most should be the easiest thing to reach.

  • Content structure

    One page, one job. Pages that try to do four do none.

  • Internal linking

    Links are how you say which pages relate to which.

  • Trust

    Real names, real proof, real answers to awkward questions.

  • Mobile

    Designed for a thumb and a distracted moment, not a desk.

  • Conversion

    One obvious action per page, phrased the way people think.

  • Speed

    The cheapest conversion win nobody wants to do.

  • Semantic clarity

    Say what the thing is, in the words people use for it.

How I think about it

A useful digital system has layers.

Most problems are not where people look for them. A conversion problem is often an information problem one layer up.

  1. 01 Discovery

    AI search · Google · Social · Marketplaces

  2. 02 Information

    Content · Schema · Reviews · Entity signals

  3. 03 Experience

    Website · Product · UX · Mobile

  4. 04 Conversion

    Lead · Purchase · Booking · Action

  5. 05 Measurement

    Analytics · Search Console · Prompt tracking · CRM

Putting it into practice

Where I apply it.

Service businesses, mostly. The same three questions, asked of very different categories.

  • Opal Dental Care logo

    Dental · Toronto + North York

    Opal Dental Care

    Procedures separated from each other, two locations treated as two real places, and the questions people ask before booking answered on the page.

    Website architecture · Procedure hierarchy · Patient journeys · Location structure

  • Maple Ridge Orthodontics logo

    Orthodontics

    Maple Ridge Orthodontics

    AI visibility and local search, starting from the least glamorous question: does the site say clearly what is offered, and where.

    AI visibility · Local search · Website clarity · Service discoverability

  • HIFA Insurance logo

    Insurance

    HIFA Insurance

    An intangible product needs more structure, not less. The work was making the services legible enough that someone could act on them.

    Website architecture · Trust · Service structure · Conversion

  • Blaze logo

    Multi-state service

    Blaze

    AEO across several states, where the hard part is coverage — one real page for every place, service and question worth answering.

    AEO · Visibility · Multi-state structure · Digital discovery

Products

Sometimes the digital foundation is the product.

Same instincts, different surface. Information architecture matters more inside an app than it does on a marketing page.

  • Coachology logo

    Coachology

    A coaching platform where the whole problem is getting a coach from sign-up to a first useful session without losing them.

    • Onboarding
    • Dashboards
    • User journeys
  • Milo Tennis logo

    Milo Tennis

    Video-led coaching, which means the hard parts are upload flows and making feedback easy to find again later.

    • Upload flows
    • Information architecture
    • Product UX
  • Coach Mahjy logo

    Coach Mahjy

    A mahjong coaching app. Teaching a complex game is an information design problem before it is a software one.

    • Learning structure
    • Onboarding
    • Product UX

Point of view

I use AI to remove friction, not add theatre.

  • Automate the repetitive work
  • Make information easier to find
  • Improve the workflow, not the demo
  • Support better decisions with real data
  • Accelerate research and drafts
  • Keep humans where judgment matters

Let's talk

Want to talk digital?

Websites, AI search, product ideas, conversion, automation, or whatever weird thing the internet is doing next.

Built, not theorized.